Understanding the state of racial/ethnic disparities in mental health care has been limited by a lack of consensus on appropriate methods. The predominant methods used in the literature are to compare unadjusted means, or to use multivariate regression methods to identify the independent association between race and mental health care. In light of the Institute of Medicine (IOM) definition of disparity in quality of health care, as presented in Unequal Treatment, these methods may over- or under-estimate disparities. The IOM defines disparities as all differences except those due to clinical appropriateness, need, and preferences. The objective of this application is to improve the identification of disparities in mental health care by refining and applying methods conceptually grounded in the IOM definition. The research will increase understanding of how the burden of mental illness differs by race/ethnicity. Progress can then be tracked towards the reduction of disparities in mental health care, with the ultimate goal of reducing the burden of mental illness felt by all racial/ethnic groups. The proposed research implements the IOM definition of racial disparities by adjusting for health status while allowing the mediation of socioeconomic status (SES) variables. Two adjustment methods are used: a rank and replace method and a propensity score-based method. These methods are refined and applied to mental health status and utilization measures in commonly used, nationally representative datasets. The specific aims of this proposal are: 1. To refine empirical methods concordant to the IOM definition of racial disparities, and apply these methods to assess the current state of black-white and Hispanic-white disparities in mental health services utilization. 2. To develop recommendations for the use of IOM-concordant methods in different data contexts. Capitalizing on the strengths of the Medical Expenditure Panel Survey (MEPS) and the National Health Interview Survey (NHIS), apply sensitivity analyses to assess the validity of these methods in calculating disparities when using different dependent variables and datasets. [unreadable] [unreadable] [unreadable]